Diabetes affects the lives of 7 million older adults aged 65+. Diabetes disproportionately affects minority older adults. Patients with diabetes are expected to take responsibility for managing their disease by practicing a set of behaviors to maintain glycemic control and minimize the occurrence of its devastating complications. Most programs to promote adequate self-management are focused on providing education and skills to patients with the intent of changing their behaviors. However, a growing literature shows that persons'approaches to disease management are grounded in common sense or self-regulatory models, complex understandings of the disease and its management that are based on life experiences and their cultural milieu. Despite the acknowledged importance of these common sense models, few attempts have been made to measure them. The long term goal of this research program is to understand how common sense models of diabetes are related to glycemic control and complications of diabetes. This project's original funding period produced an understanding of self-management behaviors and their association with glycemic control in a multi-ethnic (African American, Native American, white) population in rural North Carolina. The proposed funding period will be devoted to documenting common sense models of diabetes underlying behavior, delineating the factors related to variability in these models, and developing an inventory/instrument that measures these models in older individuals with diabetes who reside in rural and minority communities. The proposed research will use a mixed method approach (in-depth interviews, free- listing, consensus panels, and survey) to accomplish four specific aims: (1) delineate the common sense models of diabetes held by older rural adults;(2) determine the sources of variability in these common sense models, including gender and ethnicity;(3) develop the Common Sense Model Inventory (CSMI), a quantitative instrument to differentiate older adults with diabetes according to their common sense models of diabetes;and (4) establish the validity and reliability of this instrument. Relevance to public health: This study will extend knowledge of the underlying models that affect individual health behavior, and provide a tool that will allow the investigators and other researchers to test hypotheses concerning the linkage of culture through individual experience to behavior and, finally, to disease outcomes in future research.